AI and ethics最新文献

Practical wisdom (phronesis) is the context-sensitive capacity to skillfully achieve morally good outcomes in complex situations. Developing artificial practical wisdom offers a more ethically robust and achievable goal for AI development than artificial general intelligence (AGI). While identifying what is morally good in ethically complex situations remains challenging, grounding artificial practical wisdom explicitly in compassion effectively reduces ethical risks associated with AI-induced suffering, surpassing conventional alignment strategies like rule-based guardrails or predefined reward systems. As a theoretical foundation for initial development of artificial practical wisdom, this virtue ethics approach integrates Aristotelian practical wisdom with cross-cultural perspectives on suffering and compassion from utilitarianism, the Capability Approach, Buddhism, and contemporary moral psychology. Operationalizing compassionate AI involves recognizing suffering, empathetic engagement, context-sensitive moral decision making, and motivational responses. Compassionate AI not only serves as a foundation for broader practical wisdom development but also demonstrates immediate practical benefits, particularly in healthcare, by measurably improving patient outcomes, enhancing well-being, and reducing caregiver burdens.

Oral cancer remains a major yet underdiagnosed burden in low- and middle-income countries (LMICs), where high-risk behaviors, late-stage presentation, and limited diagnostic infrastructure contribute to poor outcomes. Artificial intelligence (AI), particularly vision-language foundation models such as CLIP and SAM, has emerged as a potential tool for zero-shot oral lesion detection, offering scalable diagnostic support without requiring large annotated datasets. However, these models are predominantly trained on Western-centric, light-skinned image corpora, raising concerns about fairness and generalizability in global health contexts. This review critically explores the diagnostic capabilities, limitations, and ethical considerations of using foundation models for oral lesion detection in LMICs, emphasizing their potential impact on equity and data justice. We analyzed interdisciplinary literature spanning medical imaging, oral oncology, digital health ethics, and AI fairness. Empirical studies evaluating AI performance across photographic, histological, and cytological data types were examined alongside implementation case studies and policy frameworks. Key themes included model generalization, domain adaptation, and data governance. While foundation models achieve competitive zero-shot performance in select lesion classification and segmentation tasks, their sensitivity declines when applied to darker-skinned individuals and field-acquired images from LMICs. Bias mitigation strategies such as prompt engineering, few-shot fine-tuning, and federated learning show promise but remain underutilized. Current development pipelines often lack transparency, community participation, and subgroup validation, leading to epistemic and diagnostic inequities. Foundation models offer significant potential for democratizing oral cancer screening, particularly in resource-constrained settings. However, realizing this promise requires locally grounded adaptation, inclusive dataset design, and participatory governance. Without these, zero-shot AI may reinforce rather than resolve existing disparities in oral healthcare. These findings emphasize the need for actionable collaboration among policymakers, clinicians, and AI developers. Integrating fairness metrics into regulatory review and encouraging locally validated deployment protocols can help translate these technologies into safe, equitable screening pathways in low-resource settings.

The paper explores how AI-driven truth technologies—tools and methods for determining the accuracy and truthfulness of information, statements, or claims—can be developed and utilised to mitigate misinformation and disinformation. Expectations, opportunities, risks, and governance principles are identified through 23 semi-structured qualitative interviews with distinguished professionals and academics. The results suggest that AI capabilities are expected to continue advancing and rapidly escalating AI-generated misinformation and disinformation, which undermine democracies and amplify societal polarisation. AI-driven truth technologies present opportunities to counter this development, including improved fact-checking, evaluation of publishing sources, and the contextualisation of information. Hence, truth technologies are expected to be locked into an inevitable arms race against increasingly sophisticated disinformation techniques. The results also suggest risks associated with such truth technologies, including oversimplification and bias, users’ overreliance, and misuse for manipulation or censorship, and emphasise the need for effective management and regulation to mitigate these risks. Truth technologies should, as such, be decentralised, transparent, explainable, overseen by trustworthy bodies, and complemented by education in critical thinking. In addition to the importance of effective management and public education, the paper emphasises the crucial importance of understanding the effects of power structures in shaping the development of AI. Users, developers and regulators of AI-driven truth technologies are encouraged to consider the expectations, scale the opportunities, mitigate the risks, and follow the outlined governance principles to maximise the potential of such truth technologies in counteracting misinformation and disinformation.

We examine whether large language models reproduce human-like cultural identity or instead produce deterministic cultural prototypes. Using the 30-item Self-Construal Scale (SCS), we asked two GPT models to answer as if they were American, Polish, Japanese, or with no specified identity, in English or Polish, under weak vs. strong identity cues (2 × 4 × 2; 30 runs per cell). Identity prompts yielded almost perfectly separable cultural profiles with near-zero within culture variance. In neutral conditions, outputs systematically skewed toward a U.S. profile in 75% of tests, consistent with an anglocentric default. A simple nearest-neighbor classifier achieved 99.8% leave-one-out accuracy in predicting the assigned cultural identity from item-level responses, confirming near-perfect profile separability. These findings indicate that, for cultural identity reproduction, current models behave as caricaturists rather than mirrors of human cultural variation.

This study reformulates the discussion on the integration of artificial intelligence (AI), viewing it not solely as a substitute for human labor but as a stimulus for the revival of soft skills. As AI progressively automates routine and technical tasks, it concurrently heightens the requirement for high-value human competencies including emotional intelligence, creativity, and adaptive leadership. The article examines AI-driven platforms that are transforming soft skills development via scalable and tailored learning techniques, such as virtual reality simulations and predictive analytics. The primary assertion indicates that a sustainable future of employment relies on the collaborative interplay between human and artificial intelligence. The discussion concludes by addressing essential ethical considerations, emphasizing the necessity of a human-centered framework to cultivate an inventive, responsible, and ethically conscious workforce.

Generative AI has significantly impacted creative practices, prompting a fundamental reconsideration of what constitutes creativity. Traditionally, creativity has been understood mainly as a human capacity shaped by historical and cultural contexts, and the emergence of generative AI has prompted a reexamination of the conceptual boundaries and epistemological status of creativity. This study explores how the creativity exhibited by generative AI differs from traditional human creativity, and how these systems reshape conceptual understandings of creativity and transform creative practices. Our principal contribution is a normative framework for human–AI co-creation, articulated through ethical, creative, and social principles (grounded in Beneficence/Non-Maleficence, Justice/Fairness, Autonomy, and Explicability). We operationalize these principles into stakeholder guidance, thereby yielding actionable criteria for ethical, responsible, and meaningful co-creation.

This paper examines how fairness principles differ when evaluating large language model (LLM) outputs in fact-based versus opinion-based contexts, focusing on gender disparities in responses related to notable individuals. Using prompts designed to elicit either factual information (identifying Nobel Prize winners) or subjective judgments (identifying the most accomplished figures in a field), we analyze responses from GPT-4, Claude, and Llama-3. For fact-based tasks, fairness is assessed through correctness and refusal rates, revealing minimal gender disparities when models achieve high accuracy, although refusal patterns can vary by model and gender. For opinion-based tasks, where no single correct answer exists, fairness is operationalized through representational metrics such as demographic parity and disparate impact. Results show substantial gender disparities in opinion-based outputs across all models, with representation shaped by prompt wording (e.g., “important” vs. “prestigious”), subject domain, and inclusion of secondary answers. However, the highly skewed context makes the final assessment about fairness challenging. Our findings highlight that fairness metrics and interpretations must be contextualized by output type. Performance parity is an appropriate goal for fact-based outputs, whereas representational inclusivity is central for opinion-based outputs. Representational inclusivity alone may not be sufficient when the context for the LLM’s task differs from the population. We discuss theoretical implications for fairness evaluation, noting that high performance can mitigate disparities in factual contexts but that opinion-based contexts require more nuanced, values-driven approaches.

The expanding presence of Artificial Intelligence (AI) in education raises new pedagogical possibilities as well as important ethical considerations. This study explored teachers’ perceptions and experiences related to the ethical dimensions of AI integration in the Israeli school system. Using a mixed-methods design, quantitative data were collected from 108 teachers and complemented by qualitative analyses of open-ended reflections. The quantitative results indicated strong endorsement of ethical principles, including transparency, fairness and accountability, alongside moderate levels of self reported confidence and limited participation in formal AI related training. The qualitative findings showed that while many teachers had not yet encountered AI related dilemmas directly, they described a range of possible approaches to hypothetical situations, including dialogic engagement, establishing classroom guidelines and applying corrective measures.

These findings provide a descriptive picture of how teachers currently understand AI related ethical issues in contexts where institutional norms and classroom practices are still developing. To support future discussion, the study introduces the Ethical Digital Competence (EDC) framework as a conceptual lens that highlights the interplay between knowledge, values and reflective practice. The framework is intended as a contribution to ongoing conversations about responsible AI integration and may inform the design of professional learning opportunities as educational systems continue to adapt to emerging technologies.

This study develops a stakeholder-informed artificial intelligence (AI) ethics governance checklist tailored for Canadian law firms to help them harness the productivity and economic advantages of AI while minimizing the risks of unethical outcomes. Recognizing the limitations of existing AI principles (AIPs) in preventing unethical outcomes, this research uses semi-structured interviews, qualitative content analysis, and expert stakeholder engagement to design an eight-page AI ethics governance checklist. In addition to the output of a practical governance checklist, the study reports findings about the development of stakeholder-informed governance checklists. The findings reveal that Canadian lawyers share global concerns surrounding AI risks, including privacy, accountability, safety and security, transparency and explainability, human oversight, professional responsibility, and the promotion of human values. In addition, many law firms interact with AI primarily through third-party vendors, making a principle-based checklist the most practical approach. The research highlights the importance of question format, suggesting that balancing clarity (using Yes/No options) with flexibility (allowing for open-ended comments) is essential, given the complex ethical considerations. The study also finds there is a need to integrate the checklist with existing policies, such as privacy impact assessments and IT risk evaluations, alongside relevant regulatory frameworks. Additionally, tailoring language and definitions to reflect the specific needs of stakeholders (in this case, lawyers) enhances usability and effectiveness. The resulting eight-page, stakeholder-informed AI ethics governance checklist has been adopted by several Canadian law firms and Barristers’ Societies, offering a practical tool to guide the responsible adoption and use of AI in the legal sector.

Contemporary artificial intelligence (AI) technologies, particularly those based on foundation models and released at scale, are globally entangled and made up of a complex array of interrelated actors, practices, and transnational flows of resources. The rapid pace at which AI systems are being developed and distributed, is driving significant societal and planetary transformations. While much of the international agenda around governing AI has converged around downstream matters of safe deployment and use, deeper systemic issues—including power concentration, uneven environmental costs, or the asymmetric extraction of data and labour by technology companies—remain contested and unresolved areas of debate. In this paper we centre these systemic challenges, and locate governance entry points aimed at fostering more just futures. We conceptualise contemporary large-scale AI as a sociotechnical ecology comprised of interrelated actors, practices, and asymmetrical resource flows. Using the lens of infrastructural inversion within social studies of infrastructure, we trace the actors involved in the making of AI technologies, their interdependences, and long-term infrastructural continuities that shape them. We argue that new AI models and systems are not unprecedented but are instead built upon and shaped by pre-existing infrastructures, entrenched market relations, and socio-historical patterns. By making visible the sites of accountabilities and technical and non-technical intervention in the AI ecology, we identify four governance imperatives for sustainable and equitable AI governance: (1) Decentralising AI infrastructure, (2) Advancing environmental and epistemic justice through pluriversal AI governance, (3) Instituting cross-border data and data work governance, and (4) Enhancing international coordination, participation and solidarity.